meta-feedback

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input to generate GitHub issues via shell commands. * Ingestion points: User responses to AskUserQuestion (e.g., symptom, reproduction steps, request content) in SKILL.md. * Boundary markers: Absent. There are no instructions to escape or delimit the user content before it is interpolated into the gh command. * Capability inventory: Uses the gh CLI tool to create issues on the modu-ai/moai-cowork repository. * Sanitization: Absent. The command template --body "{본문}" does not specify how to handle special characters or escape shell sequences.
  • [COMMAND_EXECUTION]: The workflow involves executing the gh issue create command with arguments derived from user input. This creates an attack surface where a user could attempt to manipulate the command parameters if the execution environment does not provide automatic escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 01:41 PM
Security Audit — agent-trust-hub — meta-feedback